1. Field of the Invention
This invention relates to a system and method for supporting operations for input of user commands to household electric appliances such as a television set/monitor and information equipment, and in particular, to an interactive operation support system and a method therefor, which permit input of user commands to various kinds of connected equipment interactively.
More specifically, the present invention is concerned with an interactive operation support system and a method therefor, which are adapted for input of user commands to the equipment in a natural form through a personified assistant, and in particular, to an interactive operation support system and a method therefor, which permit input of user commands by means of interaction with a personified assistant on a speech input base.
2. Description of the Related Art
Conventionally, an operation panel, a keyboard and a mouse or the like of a type requiring manual operation are mainly applied to a user interface, that is, input of commands to various kinds of household electric equipment such as a television set/monitor or information equipment including a personal computer. With the improvement of operation input performance of a processor and the advance of cognitive technology including speech recognition, an interactive speech-based input user interface is now widespread as well.
Since the user interface based on the manual operation permits direct and uniform input of commands to the equipment, a user may put the input operations into practice with certainty. However, the user has to understand and further get skilled in the techniques for operating the equipment to a certain extent, and therefore, an excessive burden is required for the user.
For instance, “a fingertip operation based interfaces” for control a menu with a ten-key or the like is mainly used in a television and other AV equipment. However, it is to be supposed that complicated operations are required for the interface in the user input mode described in the above to deal with network-connected household electric equipment.
While a user interface using a commander is now being generalized, too numerous switches are required to meet demands for multi-channel and multi-control including ground wave band, satellite systems, the Internet and HAVI (Home Audio/Video Interoperability: common command system for digital AV equipment), resulting in an increasing number of switches, thus making operation increasingly complicated. Combination of the above user interface with a multi-function switch and a menu screen permits a reduction in number of switches up to a certain point, however operation becomes very complex.
On the other hand, the user interface on the speech input base makes it possible to specify a command by analyzing a user request based on the result of input speech recognition on the equipment side, resulting in a relief from user's burden on the occasion of operation of the equipment. However, since it is necessary for the user to speak at a microphone in the absence of a partner, such operation hardly can be considered a naturally human action. Besides, the user may be subject to suffering mental anguish when carrying out interaction with such a kind of user interface.
In this connection, there is recently provided an interactive operation support system, which is set to allow a personified assistant to appear on a display screen, permitting the user to perform input of commands to the equipment in the form of carrying out a conversation face to face with an assistant on the screen.
In Japanese Patent Laid-open No. 11-65814, for instance, there is disclosed a user interface, which provides a sharp feeling of presence and actuality for the user by detecting a sound produced from the user or a direction of a sound source to control a visual image of an imaginary creature according to the result of detection (i.e., an imaginary creature is set to follow the source of sound by constantly gazing at the source direction).
Also, in Japanese Patent Laid-open No. 11-37766, there is disclosed an agent device, which provides a personified agent having functions of establishing communication with a driver in a vehicle. According to such agent device, the personified agent is set to make motions fit for the current conditions of the vehicle and the driver according to not only the current conditions of the vehicle and the driver but also the learning effects based on the past history, permitting the vehicle to establish communication with the driver.
Recently, an increase in computer processing power or the like permits a high-level interactive processing and also makes it possible to provide a sort of intelligence to the assistant on the screen. For instance, the assistant may not only operate an input command simple enough to be formed by a single word, such as a channel select command and a recording/reproduction start or stop command, but also perform complicated operations across a plurality of stages in pursuit of a context of the content of conversation with the user.
However, such a system making it possible to present the status conditions in progress related to such operations on the screen through an assistant has not been developed so far, and as a result, there is no other way other than the user having to wait for response from the system with one's eyes fixed on the screen. It is to be even supposed that if the user gives a command to the system to execute a processing requiring a response time, the user would even misunderstand that the equipment is out of order.
Thus, it is preferable that, in order to allow the user to operate the equipment based on interaction with the assistant, an operationally easy input of a command system produces an effect close to natural language is provided.